High-Throughput Determination of Biochemical Oxygen Demand

The use of the conventional 5-day biochemical oxygen demand (BOD5) method in BOD determination is greatly hampered by its time-consuming sampling ...
0 downloads 0 Views 211KB Size
Environ. Sci. Technol. 2007, 41, 4038-4044

High-Throughput Determination of Biochemical Oxygen Demand (BOD) by a Microplate-Based Biosensor HEI-LEUNG PANG, NGA-YAN KWOK, PAK-HO CHAN, CHI-HUNG YEUNG, WAIHUNG LO, AND KWOK-YIN WONG* Department of Applied Biology and Chemical Technology, The Hong Kong Polytechnic University, Hunghom, Kowloon, Hong Kong SAR, China

The use of the conventional 5-day biochemical oxygen demand (BOD5) method in BOD determination is greatly hampered by its time-consuming sampling procedure and its technical difficulty in the handling of a large pool of wastewater samples. Thus, it is highly desirable to develop a fast and high-throughput biosensor for BOD measurements. This paper describes the construction of a microplatebased biosensor consisting of an organically modified silica (ORMOSIL) oxygen sensing film for high-throughput determination of BOD in wastewater. The ORMOSIL oxygen sensing film was prepared by reacting tetramethoxysilane with dimethyldimethoxysilane in the presence of the oxygensensitive dye tris(4,7-diphenyl-1,10-phenanthroline)ruthenium(II) chloride. The silica composite formed a homogeneous, crack-free oxygen sensing film on polystyrene microtiter plates with high stability, and the embedded ruthenium dye interacted with the dissolved oxygen in wastewater according to the Stern-Volmer relation. The bacterium Stenotrophomonas maltophilia was loaded into the ORMOSIL/ PVA composite (deposited on the top of the oxygen sensing film) and used to metabolize the organic compounds in wastewater. This BOD biosensor was found to be able to determine the BOD values of wastewater samples within 20 min by monitoring the dissolved oxygen concentrations. Moreover, the BOD values determined by the BOD biosensor were in good agreement with those obtained by the conventional BOD5 method.

Introduction The conventional 5-day biochemical oxygen demand (BOD5) method has been routinely used to monitor organic pollutants in wastewater (1-3). This method functions by monitoring the consumption of dissolved oxygen (DO) by microorganisms (e.g., bacteria) in the oxidation of organic compounds. The BOD5 method is, however, very time-consuming (5 days) and is difficult to simultaneously screen a large pool of wastewater samples (1-4). As such, it is highly desirable to develop a fast biosensor that can detect the dissolved oxygen in wastewater samples in a high-throughput manner. Since the emergence of the first optical BOD biosensor (1), extensive efforts have been made to improve the sensing technology of BOD biosensors, ranging from remote sensing to highthroughput screening (1, 3-6). Recently, we have developed a prototype optical BOD biosensor that can simultaneously * Corresponding author phone: +852 34008686; fax: +852 23649932; e-mail: [email protected]. 4038

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 11, 2007

monitor the dissolved oxygen levels in six wastewater samples (3). Given the mature technology of microplate-based sensing, we reasoned that the sampling capacity of this optical biosensor can be significantly increased if it functions on a microplate-based platform (e.g., 96-well microtiter plate). The key to constructing microplate-based biosensors lies in the modification of chemically inert polystyrene microtiter plates with sensing materials. To this end, water-based (7) and alcohol-soluble polymers (8), which are compatible with polystyrene microtiter plates, have been developed as supporting materials for oxygen-sensitive dyes. Recently, a number of new oxygen sensors have been successfully constructed by incorporating the oxygen-sensitive dye Ru(dpp)32+ into organically modified silicates (ORMOSIL) (912). ORMOSIL is transparent, hydrophobic, photochemically stable, and highly porous (allowing fast O2 diffusion), and can provide a homogeneous environment to the embedded dyes (9-12). Moreover, ORMOSIL can be easily prepared by condensation in an aqueous medium (13), which is benign to polystyrene microtiter plates. These advantageous properties render ORMOSIL an excellent supporting matrix for the modification of polystyrene microtiter plates with oxygensensitive dyes. The Ru(dpp)32+-doped ORMOSIL film interacts with oxygen according to the Stern-Volmer relation (9-12), and is suitable for dissolved oxygen measurements (13). In this paper, we describe the construction of a microplatebased luminescent biosensor for BOD measurements. This biosensor was fabricated by combining a Ru(dpp)32+-doped ORMOSIL film with a microbial film of Stenotrophomonas maltophilia (a bacterium capable of metabolizing organic compounds (14)) in a 96-well polystyrene microtiter plate. This biosensor is easy to calibrate with high linearity over a wide range of dissolved oxygen concentrations, and can measure the BOD values of wastewater samples in a highthroughput manner using a microplate reader.

Experimental Section Materials. [Ru(dpp)3]Cl2 (dpp ) 4,7-diphenyl-1,10-phenanthroline) was synthesized and purified according to a literature method (15). Tetramethyl orthosilicate (TMOS) (98%), methyltrimethoxysilane (MTMS) (95%), dimethoxy dimethylsilane (DiMe-DMOS) (95%) and poly(vinyl alcohol) (PVA, molecular weight: 124 000-186 000) were obtained from Aldrich. Fine graphite powder (extra pure) was purchased from Merck. Nutrient agar, nutrient broth, marine broth, D-glucose, tryptone, yeast extract, peptone, beef extract, and urea were obtained from Sigma. Sterile polystyrene microtiter plates (no. 3603, 96 wells, TC-treated, flat clear bottom and black-walled) were obtained from Corning. Aluminum microplate sealing foils were purchased from USA Scientific. Milli-Q deionized water was used throughout this study. Oxygen and nitrogen gas (99.7%) were obtained from Hong Kong Oxygen Company. Wastewater samples were collected from the Shatin Sewage Treatment Works of Drainage Services Department, Hong Kong. The bacterium Stenotrophomonas maltophilia used in this study was isolated from activated sludge samples collected from the Shatin Wastewater Treatment Plant. This bacterial strain, which has high catabolic activity against simple organic compounds such as glucose-glutamic acid (14), was identified by the MIDI Sherlock Microbial Identification System in Microbial ID Inc. (Newark, DE). The standard glucose-glutamic acid (GGA) solution was prepared according to the Japanese Industrial Standard (16) in which a mixture of 150 mg l-1 glucose and 150 mg l-1 glutamic acid gives a BOD5 value of 220 ( 10 mg l-1. The glucose and glutamic acid were dried 10.1021/es070083k CCC: $37.00

 2007 American Chemical Society Published on Web 05/01/2007

in an oven at 103 °C for 1 h prior to the preparation of the standard solution. The synthetic wastewater was prepared according to the Organization for Economic Cooperation and Development (OECD) standard (17). The OECD synthetic wastewater contained peptone (1.6% w/w), meat extract (1.1% w/w), urea (0.3% w/w), sodium chloride (0.07% w/w), calcium chloride dihydrate (CaCl2‚2H2O, 0.04% w/w), magnesium sulfate heptahydrate (MgSO4‚7H2O, 0.02% w/w) and potassium monohydrogen phosphate (K2HPO4, 0.28% w/w) in 1 liter of water. Instrumentation. All luminescence measurements were performed on a BMG Labtech POLARstar microplate reader. The excitation and emission wavelengths of the bandpass filters were 460 and 610 nm, respectively. All luminescence measurements were performed at room temperature (25 °C) in the bottom-plate-reading mode unless otherwise stated. Immobilization of ORMOSIL Oxygen Sensing Films on Microtiter Plates. ORMOSIL oxygen sensing films were prepared according to the method described in our previous study (13). Silica sol was prepared by mixing TMOS with DiMe-DMOS, 0.1 M HCl, and deionized water in a ratio of 1:2.15:1.7:1.1 (v/v). This mixture was stirred for 3.5 h in a water bath (25 °C) and subsequently centrifuged at 7500 rpm for 4 min. The thickened sol at the bottom of the centrifuge tube (1 mL) was collected and thoroughly mixed with a solution of [Ru(dpp)3]Cl2 (250 µL, dissolved in ethanol at a concentration of 2.85 mg mL-1). A 32 µL portion of the sol/ ruthenium mixture was then transferred to each well of a 96-well microtiter plate and incubated for 6 days at room temperature in the dark. This volume was chosen because the sol/ruthenium mixture was found to form a film capable of completely covering the bottom of the well (13). The thickness of the central part of the film was estimated to be less than 10 µm based on the amount of the sol/dye mixture applied to the well (13). A graphite-doped solution of solgel was prepared by mixing 1 mL of the thickened sol with 0.1 g of graphite and 50 µL of 0.1 M phosphate buffer (pH 7.0). This mixture (15 µL) was then coated onto the gelatinized oxygen sensing films so as to reduce the scattered light and background luminescence from samples. The film-coated microtiter plate was left in the dark at room temperature for another 6 days. The oxygen-sensing ability of the Ru(dpp)32+doped film was investigated by luminescence measurements with different concentrations of oxygen in deionized water. Preparation of Stenotrophomonas maltophilia. A single colony of Stenotrophomonas maltophilia was inoculated into 100 mL of sterile medium containing 1.5% (w/v) peptone, 1.2% (w/v) sodium chloride, 0.5% (w/v) glucose, 1% (w/v) tryptone, 0.3% (w/v) beef extract, and 1.9% (w/v) yeast extract in a 1-liter incubation flask. The culture was incubated for about 24 h at 30 °C with shaking at 250 rpm, followed by centrifugation at 6000 rpm for 5 min. The cell pellet was collected and washed twice with 0.1 M phosphate buffer (pH 7.5). The resulting cell pellet was resuspended in 0.1 M phosphate buffer (pH 7.5) as the bacterial stock. Construction of the BOD Biosensor. The BOD biosensor was constructed by coating a microbial film onto the oxygen sensing film as follows. A stock solution of sol-gel was first prepared by mixing 0.74 mL of TMOS with 0.36 mL of MTMS and 0.5 mL of 0.01 M HCl, followed by stirring at room temperature for half an hour. This mixture was then mixed with an aqueous solution of PVA (8%, w/w) in a ratio of 2:1 (v/v). Microbial films with different amounts of bacteria were prepared by adding the bacterial stock (0.6-1.8 mg mL-1) to the silica sol/PVA solution, followed by the coating of these mixtures (20 µL) onto the graphite-containing sol-gel films. This mixture was left at 4 °C for 24 h for gelation. The bacteriacoated sol-gel films were then soaked with a phosphatebuffered solution of GGA (100 mg l-1, pH 7.5) at 4 °C before

FIGURE 1. Schematic diagram of the BOD biosensor. use. A schematic diagram of the BOD biosensor is shown in Figure 1. Monitoring of Dissolved Oxygen. The BOD biosensor (without the microbial film) was calibrated for DO detection according to the method described in our previous study (13). Briefly, nitrogen and oxygen were mixed together in various volume ratios in a glass chamber, using gas flowmeters. The outlet of the glass chamber was connected to a tubing, which was immersed into deionized water. A series of aqueous solutions with different concentrations of dissolved oxygen was prepared by pumping the water samples with the N2/O2 mixtures for 30 min, and the DO concentrations of these samples were determined using a YSI 5000 dissolved oxygen meter (YSI Inc., Yellow Springs, Ohio). For the O2-free water sample, sodium sulfite (0.5 M) and cobalt sulfate (0.1 µM) were used to remove the dissolved oxygen. These samples were then transferred to the wells immediately, and the microtiter plate was sealed with an aluminum microplate sealing foil to prevent atmospheric oxygen from dissolving into the samples. The luminescence signals of these samples were recorded and used to construct a calibration plot of Io/I versus DO concentration, where Io and I represent the luminescence intensities recorded in the absence and presence of O2, respectively. Unless otherwise stated, all the 96 wells of the BOD biosensor were used in DO measurements. For the calibration of the BOD biosensor (with the microbial film), luminescence measurements were performed with the BOD biosensor in the presence of different concentrations of GGA (0-110 mg L-1). A 250 µL portion of each standard solution was added to the well of the BOD biosensor and the microtiter plate sealed with an aluminum sealing foil. This sample volume was chosen because the well of the BOD biosensor was almost fully occupied. The luminescence signals of the standard solutions were recorded over the time course, which were subsequently converted into DO concentrations using the calibration plot of Io/I versus DO concentration (see above). These data were used to construct a plot of DO concentration versus time from which the initial DO consumption rates (in the first 20 min) for the standard solutions were determined. A calibration plot of VOL. 41, NO. 11, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4039

initial DO consumption rate versus GGA concentration was then constructed with these data. Similar luminescent measurements and data analysis were also performed with domestic wastewater samples and OECD. For comparison, the conventional 5-day BOD (BOD5) values of the GGA, OECD and domestic wastewater samples were determined by the standard method (18). Cell Loading. The luminescence signals of the BOD biosensor loaded with different amounts of Stenotrophomonas maltophilia (0.6-1.8 mg mL-1) with GGA (BOD5: 20 mg l-1) were measured (at 25 °C). The resulting luminescence profiles were then analyzed to determine the initial DO consumption rates (in the first 20 min) using the method described above. Effects of pH and Temperature. The effect of varying pH on the catabolic activity of the BOD biosensor was studied by monitoring catabolic process of the GGA samples (with a BOD5 value of 20 mg L-1) at different pH values (6.0-9.0) through luminescence measurements (at 25 °C). The initial DO consumption rate (in the first 20 min) for each sample was then determined by analyzing the luminescence profile using the method described in the previous section (Monitoring of Dissolved Oxygen). The effect of varying temperature on the catabolic activity of the BOD biosensor was examined by similar strategies except that the experiments were performed at pH 7.5 with the temperature varied from 25 to 45 °C. Effects of Heavy Metal Ions. The catabolic activity of the BOD biosensor in the presence of each of the metal ions (3.0 ppm Zn2+, Pb2+, Cu2+, Ni2+, Cd2+, and Cr3+) was studied by monitoring the catabolic process of the GGA sample (with a BOD5 value of 20 mg L-1) at pH 7.5 through luminescence measurements. The luminescence profile for each sample was then analyzed to determine the initial DO consumption rate (in the first 20 min) using the method described in the previous section (Monitoring of Dissolved Oxygen). For comparison, similar luminescence measurements were also performed with the BOD biosensor in the absence of the metal ions. Stability and Service Life of the BOD Biosensor. The BOD biosensor was stored in phosphate buffer (pH 7.5) with GGA (100 mg L-1) at 4 °C for 1 month. The catabolic activities of the BOD biosensor on different days of storage were examined by monitoring the catabolic process of the GGA sample (with a BOD5 of 20 mg L-1, pH 7.5) at 25 °C through luminescence measurements. The initial DO consumption rates (in the first 20 min) were then determined by analyzing the luminescence profiles using the method described in the previous section (Monitoring of Dissolved Oxygen).

Results and Discussion ORMOSIL Oxygen Sensing Films. Many sol-gel-derived materials are unable to adhere to hydrophobic polystyrene microtiter plates due to their hydrophilic nature (13). Interestingly, our recent study has shown that ORMOSIL oxygen sensing films (doped with Ru(dpp)32+) derived from DiMe-DMOS/TMOS (13) can adhere strongly to polystyrene microtiter plates (subjected to corona discharge), and interacts with the dissolved oxygen in water according to the Stern-Volmer relation (eq 1) (19):

I0/I ) 1 + Ksv [O2]

(1)

where I0 is the luminescence intensity in the absence of oxygen, I is the luminescence intensity in the presence of O2, [O2] is the concentration of oxygen, and Ksv is the SternVolmer constant. To investigate the oxygen-sensing ability of the ORMOSIL film of the BOD biosensor (without the microbial film), luminescence measurements were performed with different 4040

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 11, 2007

FIGURE 2. Stern-Volmer plot of the ORMOSIL oxygen sensing film of the BOD biosensor. Io and I represent the luminescence intensities measured in the absence and presence of oxygen, respectively. Excitation wavelength: 460 nm; emission wavelength: 610 nm. Temperature: 25 °C. The error bars represent standard deviations (n ) eight measurements).

FIGURE 3. Initial dissolved oxygen consumption rates of the BOD biosensor containing different amounts of Stenotrophomonas maltophilia with 20 mg L-1 GGA in the first 20 min Excitation wavelength: 460 nm; emission wavelength: 610 nm. Temperature: 25 °C. The error bars represent standard deviations (n ) eight measurements). concentrations of O2 in deionized water (see Experimental Section). As shown in Figure 2, the luminescence intensity (I) of Ru(dpp)32+ in the oxygen sensing film decreases when the DO concentration is increased, indicating that the luminescence of this dye is quenched by the dissolved oxygen. The linear relationship of the Io/I values of the ruthenium complex with the DO concentrations (0-40 mg L-1) implies that the oxygen sensing film interacts with the dissolved oxygen according to the Stern-Volmer relation. This photophysical property, therefore, allows the BOD sensor to be calibrated through the simple two-point-calibration method, which has been applied in many commercially available polymer-based oxygen sensors (8, 20-22). The luminescence properties of the ruthenium dye in the ORMOSIL film were also studied. The variation in luminescence intensity of the BOD biosensor due to nonuniform film thickness was found to be about (3% (13). The quenching time (from deoxygenated water to 100% oxygenated water) and the recovery time (from 100% oxygenated water to deoxygenated water) of the BOD biosensor, which correspond

FIGURE 4. Time course of the dissolved oxygen concentrations in different GGA samples (0-110 mg L-1) probed by the BOD biosensor. The inset shows the change in dissolved oxygen concentration in the first 20 min in each sample.

FIGURE 5. Calibration plot for the BOD biosensor consisting of Stenotrophomonas maltophilia with different concentrations of GGA. The initial dissolved oxygen consumption rates were determined by analyzing the changes in dissolved oxygen concentration of the samples in the first 20 min. Excitation wavelength: 460 nm; emission wavelength: 610 nm. Temperature: 25 °C. The error bars represent standard deviations (n ) eight measurements). to 90% of the full signal (t90), were found to be 1.1 ( 0.1 and 2.1 ( 0.4 min, respectively (13). The effects of some organic and inorganic compounds such as phosphate, oxalate, phthalate, and EDTA (which are commonly present in wastewater (23)) and multi-metal ions (e.g., Pb2+, Cd2+, Cu2+, Cr3+, Zn2+, and Ni2+) on the luminescence of the oxygen sensor without the microbial film were also studied. As shown in Figure S1, the oxygen sensor shows no significant changes in luminescence intensity with different concentrations of these substances (0-30 ppm), indicating that the luminescence of the ruthenium dye in the ORMOSIL layer is not affected by these compounds or ions. This can be attributed to the hydrophobicity of the ORMOSIL sensing layer, which prevents the water-soluble compounds and ions from entering and hence quenching the ruthenium dye.

FIGURE 6. Initial dissolved oxygen consumption rates of the BOD biosensor with 20 mg L-1 GGA in the first 20 min at different pH values. Excitation wavelength: 460 nm; emission wavelength: 610 nm. Temperature: 25 °C. The error bars represent standard deviations (n ) eight measurements). Microbial Films. The catabolic activities of the BOD biosensor loaded with different amounts of Stenotrophomonas maltophilia were examined through luminescence measurements. The luminescence profiles of the BOD biosensor resulting from the oxidation of GGA (with a BOD5 value of 20 mg L-1) were analyzed to determine the initial DO consumption rates (see Experimental Section). Figure 3 shows the initial DO consumption rates of the BOD biosensor loaded with 0.6-1.8 mg mL-1 bacteria. In all cases, oxygen depletion is observed, indicating that the immobilized bacteria are able to metabolize GGA (Figure 3). The oxygen consumption process is significantly enhanced when the cell loading is increased to 1.4 mg mL-1. Further increasing the cell loading, however, shows no significant increase in initial DO consumption rate, presumably due to reduced bacterial activity and the limited amount of the organic substrate in the sample (3, 6, 24) (Figure 3). To ensure that the BOD biosensor gives a maximum response, the BOD biosensor VOL. 41, NO. 11, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4041

TABLE 2. Comparison of the BOD Values Obtained by the BOD Biosensor and the Conventional BOD5 Method sample

BOD biosensorc (mg L-1)

BOD5 (mg L-1)

GGAa 169.0 ( 21.9 157.4 ( 30.8 OECDb 1246.5 ( 21.8 1423.3 ( 108.8 GGA + OECD 658.2 ( 76.7 727.1 ( 77.5 domestic 154.7 ( 25.5 182.5 ( 46.9 wastewater

ratio (BOD biosensor/BOD5) 1.07 0.88 0.91 0.85

a GGA: glucose and glutamic acid (150 mg L-1 each). b OECD synthetic wastewater contains 0.16% (w/v) peptone, 0.11% (w/v) meat extract, 0.03% (w/v) urea, 0.007% (w/v) NaCl, 0.004% (w/v) CaCl2‚2H2O, 0.002% (w/v) MgSO4‚7H2O, and 0.028% (w/v) K2HPO4. c BOD biosensor ) BOD value (refers to as the GGA concentration in the BOD biosensor calibration plot, Figure 5) × dilution factor.

FIGURE 7. Initial dissolved oxygen consumption rates of the BOD biosensor with GGA in the first 20 min (20 mg L-1, pH 7.5) at different temperatures. Excitation wavelength: 460 nm; emission wavelength: 610 nm. The error bars represent standard deviations (n ) eight measurements).

TABLE 1. Inhibitory Effects of Various Heavy Metal Ions on the Catabolic Activity of the BOD Biosensor metal ion

inhibition (%)a

Zn2+ Pb2+ Cu2+ Ni2+ Cd2+ Cr3+

11.9 ( 3.6 19.2 ( 3.3 17.5 ( 7.9 6.3 ( 2.5 17.6 ( 2.2 16.3 ( 2.8

a Inhibition (%) ) [(R - R )/R ] × 100%, where R and R represent c t c c t the DO consumption rate of Stenotrophomonas maltophilia in the absence of heavy metal ions and in the presence of heavy metal ions, respectively.

with a cell loading of 1.6 mg mL-1 Stenotrophomonas maltophilia was used for further experiments (Figure 3). Calibration of the BOD Biosensor. Oxygen is consumed by bacteria in the oxidation of organic compounds. Thus, DO consumption rate represents a good indicator of the concentration of organic compounds in wastewater. To investigate the relation of the DO consumption rate of the bacteria-coated BOD biosensor with the concentration of organic compounds (e.g., GGA), luminescence measurements were performed with different concentrations of GGA. The resulting luminescence profiles were then analyzed to determine the DO concentrations using the calibration plot of Io/I versus DO concentration (see Experimental Section). Figure 4 shows the time course of the concentrations of dissolved oxygen with different concentrations of GGA probed by the BOD biosensor. The DO consumption rate increases with the concentration of GGA (0-80 mg L-1), indicating that the catabolic process is dependent on the concentration of the organic substrate under these conditions. The initial rate of DO consumption in the first 20 min in each case was then analyzed, and a calibration curve constructed by plotting the initial DO consumption rates against the BOD values of the GGA solutions (Figure 5). The BOD biosensor exhibits a linear calibration curve in the BOD concentration range of 0-70 mg L-1 with a detection limit of 5 mg L-1. This calibration curve shows a much wider working range (0-70 mg L-1) compared to that constructed by analyzing the DO consumption rates between 20 and 30 min (working range ) 0-10 mg L-1) (Figure S2). Thus, the former calibration method was adopted in all BOD measurements in this study. It is interesting to note that the working range of the BOD biosensor (0-70 mg L-1) is much wider than those of other 4042

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 11, 2007

FIGURE 8. Stability of the BOD biosensor incubated with GGA (100 mg L-1) at 4 °C and pH 7.5 over 30 days. The catabolic activity of the microbial film of the BOD biosensor was examined by monitoring the initial dissolved oxygen consumption rate in the first 20 min with GGA (20 mg L-1) at 25 °C and pH 7.5. Excitation wavelength: 460 nm; emission wavelength: 610 nm. The error bars represent standard deviations (n ) eight measurements). BOD biosensors composed of Trichosporon cutaneum ( Cu2+ > Cr3+ > Zn2+ > Ni2+ Comparison of the BOD Biosensor with the Conventional BOD5 Method. The BOD values for GGA, synthetic wastewater (OECD), and domestic wastewater determined by the BOD biosensor were compared to those obtained by the conventional BOD5 method. The BOD values for GGA, OECD, and domestic wastewater determined by both methods are very similar (Table 2) and show a good correlation (Figure S3). Moreover, the reproducibility of the response of the BOD biosensor is, in general, better than that of the conventional BOD5 method (Table 2). More importantly, the BOD biosensor requires a much shorter time to determine BOD compared to the conventional 5-day BOD method; the BOD biosensor requires 20 min only to determine DO consumption rates (and hence BOD). These observations indicate that the BOD biosensor is a reliable tool for rapid BOD measurements. Stability and Service Life. The stability of the microbial film of the BOD biosensor in the presence of GGA (100 mg L-1) at pH 7.5 and 4 °C within 1 month was studied. Briefly, the catabolic activities of the BOD biosensor on different days of storage were examined by monitoring the initial DO consumption rates for GGA (with a BOD5 value of 20 mg L-1) at 25 °C and pH 7.5 (see Experimental Section). As shown in Figure 8, the initial DO consumption rate (and hence the catabolic activity) of the microbial film decreases over the time course, presumably due to cell lysis. The BOD biosensor was found to retain about 89 and 46% of its initial catabolic activity after 8 and 30 days of incubation, respectively. Comparing with the service lives of other microbes, the service life of Stenotrophomonas maltophilia (46% after 30 days) is slightly longer than those of Pseudomonas fluorescens bv.V (34% after 30 days) (24) and mixed bacterial culture (Citrobacter sp. + Enterobacter sp.) (33% after 20 days) (32). These observations imply that the service life of the BOD biosensor can be conserved by storing it in buffered GGA (pH 7.5) at 4 °C.

Acknowledgments We acknowledge support from the Hong Kong Polytechnic University and the Research Grants Council. The Drainage Services Department of Hong Kong is thanked for providing the activated sludge and wastewater samples.

Note Added After ASAP Publication This paper was published ASAP May 1, 2007 with an incorrect Figure 5; the corrected version was published ASAP May 3, 2007.

Supporting Information Available Experimental details and figures for the studies of the effects of organic and inorganic compounds, the BOD biosensor calibration plot constructed by analyzing DO consumption rate between 20 and 30 min, and the correlation plot of BOD of the BOD biosensor and the BOD5 method. This material is available free of charge via the Internet at http:// pubs.acs.org.

Literature Cited (1) Preininger, C.; Klimant, I.; Wolfbeis, O. S. Optical fiber sensor for biological oxygen demand. Anal. Chem. 1994, 66 (11), 18411846. (2) Liu, J.; Mattiasson, B. Microbial BOD sensors for wastewater analysis. Water Res. 2002, 36 (15), 3786-3802. (3) Kwok, N. Y.; Dong, S.; Lo, W.; Wong, K. Y. An optical biosensor for multi-sample determination of biochemical oxygen demand (BOD). Sens. Actuators B 2005, 110 (2), 289-298. (4) Dai, Y. J.; Lin, L.; Li, P. W.; Chen, X.; Wang, X. R.; Wong, K. Y. Comparison of BOD optical fiber biosensors based on different microorganisms immobilized in ormosil matrixes. Int. J. Environ. Anal. Chem. 2004, 84 (8), 607-617. (5) Li, X. M.; Ruan, F. C.; Ng, W. Y.; Wong, K. Y. Scanning optical sensor for the measurement of dissolved oxygen and BOD. Sens. Actuators B 1994, 21 (2), 143-149. (6) Chee, G. J.; Nomura, Y.; Ikebukuro, K.; Karube, I. Optical fiber biosensor for the determination of low biochemical oxygen demand. Biosens. Bioelectron. 2000, 15 (7-8), 371-376. (7) Pitner, J. B.; Hemperly, J. J.; Guarino, R. D.; Wodnicka, M.; Stitt, D. T.; Burrell, G. J.; Foley, T. G. J.; Beaty, P. S. Device for Monitoring Cells. U.S. Patent 6395506, 2002. (8) John, G. T.; Klimant, I.; Wittmann, C.; Heinzle, E. Integrated optical sensing of dissolved oxygen in microtiter plates: A novel tool for microbial cultivation. Biotechnol. Bioeng. 2003, 81 (7), 829-836. (9) Chen, X.; Zhong, Z. M.; Li, Z.; Jiang, Y. Q.; Wang, X. R.; Wong, K. Y. Characterization of ormosil film for dissolved oxygensensing. Sens. Actuators B 2002, 87 (2), 233-238. (10) Koo, Y. E. L.; Cao, Y. F.; Kopelman, R.; Koo, S. M.; Brasuel, M.; Philbert, M. A. Real-time measurements of dissolved oxygen inside live cells by organically modified silicate fluorescent nanosensors. Anal. Chem. 2004, 76 (9), 2498-2505. (11) Tang, Y.; Tehan, E. C.; Tao, Z. Y.; Bright, F. V. Sol-gel-derived sensor materials that yield linear calibration plots, high sensitivity, and long-term stability. Anal. Chem. 2003, 75 (10), 24072413. (12) Bukowski, R. M.; Ciriminna, R.; Pagliaro, M.; Bright, F. V. Highperformance quenchometric oxygen sensors based on fluorinated xerogels doped with [Ru(dpp)3)]2+. Anal. Chem. 2005, 77 (8), 2670-2672. (13) Pang, H. L.; Kwok, N. Y.; Chow, M. C. L.; Yeung, C. H.; Wong, K. Y.; Chen, X.; Wang, X. ORMOSIL oxygen sensors on polystyrene microplate for dissolved oxygen measurement. Sens. Actuators B 2006, doi:10.1016/j.snb.2006.07.035. (14) Kwok, N. Y. Development of high-throughput biosensors for multi-sample determination of biochemical oxygen demand (BOD) based on optical detection of oxygen. Ph.D. Thesis, The Hong Kong Polytechnic University, Hong Kong, 2005. (15) Lin, C. T.; Bottcher, W.; Chou, M.; Creutz, C.; Sutin, N. Mechanism of quenching of emission of substituted polypyridineruthenium(II) complexes by iron(III), chromium(III), and europium(III) ions. J Am. Chem. Soc. 1976, 98 (21), 6536-6544. (16) Japanese Industrial Standard Committee. Apparatus for the Estimation of Biochemical Oxygen Demand (BOD) with Microbial Sensor; Japanese Standards Association: Tokyo, 1990; Vol. JIS K 3602. (17) Organization for Economic Cooperation and Development, Activated sludge, respiration inhibition test. OECD Guideline for Testing of Chemicals 1981, 209, 1-10. (18) APHA; AWWA; WEF, Standard Methods for the Examination of Water and Wastewater, 19th ed.; American Public Health Association: Washington, DC, 1995. (19) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 2nd ed.; Kluwer Academic/Plenum: New York, 1999. VOL. 41, NO. 11, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4043

(20) Guarino, R. D.; Dike, L. E.; Haq, T. A.; Rowley, J. A.; Pitner, J. B.; Timmins, M. R. Method for determining oxygen consumption rates of static cultures from microplate measurements of pericellular dissolved oxygen concentration. Biotechnol. Bioeng. 2004, 86 (7), 775-787. (21) Deshpande, R. R.; Heinzle, E. On-line oxygen uptake rate and culture viability measurement of animal cell culture using microplates with integrated oxygen sensors. Biotechnol. Lett. 2004, 26 (9), 763-767. (22) Hutter, B.; John, G. T. Evaluation of OxoPlate for real-time assessment of antibacterial activities. Curr. Microbiol. 2004, 48 (1), 57-61. (23) Dao, T. H. Organic ligand effects on enzymatic dephosphorylation of myo-inositol hexakis dihydrogenphosphate in dairy wastewater. J. Environ. Qual. 2004, 33 (1), 349-357. (24) Yoshida, N.; McNiven, S. J.; Yoshida, A.; Morita, T.; Nakamura, H.; Karube, I. A compact optical system for multi-determination of biochemical oxygen demand using disposable strips. Field Anal. Chem. Technol. 2001, 5 (5), 222-227. (25) Hikuma, M.; Suzuki, H.; Yasuda, T.; Karube, I.; Suzuki, S. Amperometric estimation of BOD by using living immobilized yeasts. Eur. J. Appl. Microbiol. 1979, 8 (4), 289-297. (26) Tan, T. C.; Wu, C. H. BOD sensors using multi-species living or thermally killed cells of a BODSEED microbial culture. Sens. Actuators B 1999, 54 (3), 252-260.

4044

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 11, 2007

(27) Li, Y. R.; Chu, J. Study of BOD microbial sensors for waste-water treatment control. Appl. Biochem. Biotechnol. 1991, 28/29, 855863. (28) Ohki, A.; Shinohara, K.; Ito, O.; Naka, K.; Maeda, S.; Sato, T.; Akano, H.; Kato, N.; Kawamura, Y. A BOD sensor using Klebsiella oxytoca As1. Int. J. Environ. Anal. Chem. 1994, 56 (4), 261269. (29) Li, F.; Tan, T. C. Effect of heavy metal ions on the efficacy of a mixed Bacilli BOD sensor. Biosens. Bioelectron. 1994, 9 (4-5), 315-324. (30) Wong, K. Y.; Zhang, M. Q.; Li, X. M.; Lo, W. H. A luminescencebased scanning respirometer for heavy metal toxicity monitoring. Biosens. Bioelectron. 1997, 12 (2), 125-133. (31) Chee, G. J.; Nomura, Y.; Karube, I. Biosensor for the estimation of low biochemical oxygen demand. Anal. Chim. Acta. 1999, 379 (1-2), 185-191. (32) Galindo, E.; Garcia, J. L.; Torres, L. G.; Quintero, R. Characterization of microbial membranes used for the estimation of biochemical oxygen demand with a biosensor. Biotechnol. Technol. 1992, 6 (5), 399-404.

Received for review January 12, 2007. Revised manuscript received March 13, 2007. Accepted March 30, 2007. ES070083K